1#
2# Secret Labs' Regular Expression Engine
3#
4# re-compatible interface for the sre matching engine
5#
6# Copyright (c) 1998-2001 by Secret Labs AB.  All rights reserved.
7#
8# This version of the SRE library can be redistributed under CNRI's
9# Python 1.6 license.  For any other use, please contact Secret Labs
10# AB (info@pythonware.com).
11#
12# Portions of this engine have been developed in cooperation with
13# CNRI.  Hewlett-Packard provided funding for 1.6 integration and
14# other compatibility work.
15#
16
17r"""Support for regular expressions (RE).
18
19This module provides regular expression matching operations similar to
20those found in Perl.  It supports both 8-bit and Unicode strings; both
21the pattern and the strings being processed can contain null bytes and
22characters outside the US ASCII range.
23
24Regular expressions can contain both special and ordinary characters.
25Most ordinary characters, like "A", "a", or "0", are the simplest
26regular expressions; they simply match themselves.  You can
27concatenate ordinary characters, so last matches the string 'last'.
28
29The special characters are:
30    "."      Matches any character except a newline.
31    "^"      Matches the start of the string.
32    "$"      Matches the end of the string or just before the newline at
33             the end of the string.
34    "*"      Matches 0 or more (greedy) repetitions of the preceding RE.
35             Greedy means that it will match as many repetitions as possible.
36    "+"      Matches 1 or more (greedy) repetitions of the preceding RE.
37    "?"      Matches 0 or 1 (greedy) of the preceding RE.
38    *?,+?,?? Non-greedy versions of the previous three special characters.
39    {m,n}    Matches from m to n repetitions of the preceding RE.
40    {m,n}?   Non-greedy version of the above.
41    "\\"     Either escapes special characters or signals a special sequence.
42    []       Indicates a set of characters.
43             A "^" as the first character indicates a complementing set.
44    "|"      A|B, creates an RE that will match either A or B.
45    (...)    Matches the RE inside the parentheses.
46             The contents can be retrieved or matched later in the string.
47    (?aiLmsux) Set the A, I, L, M, S, U, or X flag for the RE (see below).
48    (?:...)  Non-grouping version of regular parentheses.
49    (?P<name>...) The substring matched by the group is accessible by name.
50    (?P=name)     Matches the text matched earlier by the group named name.
51    (?#...)  A comment; ignored.
52    (?=...)  Matches if ... matches next, but doesn't consume the string.
53    (?!...)  Matches if ... doesn't match next.
54    (?<=...) Matches if preceded by ... (must be fixed length).
55    (?<!...) Matches if not preceded by ... (must be fixed length).
56    (?(id/name)yes|no) Matches yes pattern if the group with id/name matched,
57                       the (optional) no pattern otherwise.
58
59The special sequences consist of "\\" and a character from the list
60below.  If the ordinary character is not on the list, then the
61resulting RE will match the second character.
62    \number  Matches the contents of the group of the same number.
63    \A       Matches only at the start of the string.
64    \Z       Matches only at the end of the string.
65    \b       Matches the empty string, but only at the start or end of a word.
66    \B       Matches the empty string, but not at the start or end of a word.
67    \d       Matches any decimal digit; equivalent to the set [0-9] in
68             bytes patterns or string patterns with the ASCII flag.
69             In string patterns without the ASCII flag, it will match the whole
70             range of Unicode digits.
71    \D       Matches any non-digit character; equivalent to [^\d].
72    \s       Matches any whitespace character; equivalent to [ \t\n\r\f\v] in
73             bytes patterns or string patterns with the ASCII flag.
74             In string patterns without the ASCII flag, it will match the whole
75             range of Unicode whitespace characters.
76    \S       Matches any non-whitespace character; equivalent to [^\s].
77    \w       Matches any alphanumeric character; equivalent to [a-zA-Z0-9_]
78             in bytes patterns or string patterns with the ASCII flag.
79             In string patterns without the ASCII flag, it will match the
80             range of Unicode alphanumeric characters (letters plus digits
81             plus underscore).
82             With LOCALE, it will match the set [0-9_] plus characters defined
83             as letters for the current locale.
84    \W       Matches the complement of \w.
85    \\       Matches a literal backslash.
86
87This module exports the following functions:
88    match     Match a regular expression pattern to the beginning of a string.
89    fullmatch Match a regular expression pattern to all of a string.
90    search    Search a string for the presence of a pattern.
91    sub       Substitute occurrences of a pattern found in a string.
92    subn      Same as sub, but also return the number of substitutions made.
93    split     Split a string by the occurrences of a pattern.
94    findall   Find all occurrences of a pattern in a string.
95    finditer  Return an iterator yielding a match object for each match.
96    compile   Compile a pattern into a RegexObject.
97    purge     Clear the regular expression cache.
98    escape    Backslash all non-alphanumerics in a string.
99
100Some of the functions in this module takes flags as optional parameters:
101    A  ASCII       For string patterns, make \w, \W, \b, \B, \d, \D
102                   match the corresponding ASCII character categories
103                   (rather than the whole Unicode categories, which is the
104                   default).
105                   For bytes patterns, this flag is the only available
106                   behaviour and needn't be specified.
107    I  IGNORECASE  Perform case-insensitive matching.
108    L  LOCALE      Make \w, \W, \b, \B, dependent on the current locale.
109    M  MULTILINE   "^" matches the beginning of lines (after a newline)
110                   as well as the string.
111                   "$" matches the end of lines (before a newline) as well
112                   as the end of the string.
113    S  DOTALL      "." matches any character at all, including the newline.
114    X  VERBOSE     Ignore whitespace and comments for nicer looking RE's.
115    U  UNICODE     For compatibility only. Ignored for string patterns (it
116                   is the default), and forbidden for bytes patterns.
117
118This module also defines an exception 'error'.
119
120"""
121
122import enum
123import sre_compile
124import sre_parse
125import functools
126try:
127    import _locale
128except ImportError:
129    _locale = None
130
131# public symbols
132__all__ = [
133    "match", "fullmatch", "search", "sub", "subn", "split",
134    "findall", "finditer", "compile", "purge", "template", "escape",
135    "error", "A", "I", "L", "M", "S", "X", "U",
136    "ASCII", "IGNORECASE", "LOCALE", "MULTILINE", "DOTALL", "VERBOSE",
137    "UNICODE",
138]
139
140__version__ = "2.2.1"
141
142class RegexFlag(enum.IntFlag):
143    ASCII = sre_compile.SRE_FLAG_ASCII # assume ascii "locale"
144    IGNORECASE = sre_compile.SRE_FLAG_IGNORECASE # ignore case
145    LOCALE = sre_compile.SRE_FLAG_LOCALE # assume current 8-bit locale
146    UNICODE = sre_compile.SRE_FLAG_UNICODE # assume unicode "locale"
147    MULTILINE = sre_compile.SRE_FLAG_MULTILINE # make anchors look for newline
148    DOTALL = sre_compile.SRE_FLAG_DOTALL # make dot match newline
149    VERBOSE = sre_compile.SRE_FLAG_VERBOSE # ignore whitespace and comments
150    A = ASCII
151    I = IGNORECASE
152    L = LOCALE
153    U = UNICODE
154    M = MULTILINE
155    S = DOTALL
156    X = VERBOSE
157    # sre extensions (experimental, don't rely on these)
158    TEMPLATE = sre_compile.SRE_FLAG_TEMPLATE # disable backtracking
159    T = TEMPLATE
160    DEBUG = sre_compile.SRE_FLAG_DEBUG # dump pattern after compilation
161globals().update(RegexFlag.__members__)
162
163# sre exception
164error = sre_compile.error
165
166# --------------------------------------------------------------------
167# public interface
168
169def match(pattern, string, flags=0):
170    """Try to apply the pattern at the start of the string, returning
171    a match object, or None if no match was found."""
172    return _compile(pattern, flags).match(string)
173
174def fullmatch(pattern, string, flags=0):
175    """Try to apply the pattern to all of the string, returning
176    a match object, or None if no match was found."""
177    return _compile(pattern, flags).fullmatch(string)
178
179def search(pattern, string, flags=0):
180    """Scan through string looking for a match to the pattern, returning
181    a match object, or None if no match was found."""
182    return _compile(pattern, flags).search(string)
183
184def sub(pattern, repl, string, count=0, flags=0):
185    """Return the string obtained by replacing the leftmost
186    non-overlapping occurrences of the pattern in string by the
187    replacement repl.  repl can be either a string or a callable;
188    if a string, backslash escapes in it are processed.  If it is
189    a callable, it's passed the match object and must return
190    a replacement string to be used."""
191    return _compile(pattern, flags).sub(repl, string, count)
192
193def subn(pattern, repl, string, count=0, flags=0):
194    """Return a 2-tuple containing (new_string, number).
195    new_string is the string obtained by replacing the leftmost
196    non-overlapping occurrences of the pattern in the source
197    string by the replacement repl.  number is the number of
198    substitutions that were made. repl can be either a string or a
199    callable; if a string, backslash escapes in it are processed.
200    If it is a callable, it's passed the match object and must
201    return a replacement string to be used."""
202    return _compile(pattern, flags).subn(repl, string, count)
203
204def split(pattern, string, maxsplit=0, flags=0):
205    """Split the source string by the occurrences of the pattern,
206    returning a list containing the resulting substrings.  If
207    capturing parentheses are used in pattern, then the text of all
208    groups in the pattern are also returned as part of the resulting
209    list.  If maxsplit is nonzero, at most maxsplit splits occur,
210    and the remainder of the string is returned as the final element
211    of the list."""
212    return _compile(pattern, flags).split(string, maxsplit)
213
214def findall(pattern, string, flags=0):
215    """Return a list of all non-overlapping matches in the string.
216
217    If one or more capturing groups are present in the pattern, return
218    a list of groups; this will be a list of tuples if the pattern
219    has more than one group.
220
221    Empty matches are included in the result."""
222    return _compile(pattern, flags).findall(string)
223
224def finditer(pattern, string, flags=0):
225    """Return an iterator over all non-overlapping matches in the
226    string.  For each match, the iterator returns a match object.
227
228    Empty matches are included in the result."""
229    return _compile(pattern, flags).finditer(string)
230
231def compile(pattern, flags=0):
232    "Compile a regular expression pattern, returning a pattern object."
233    return _compile(pattern, flags)
234
235def purge():
236    "Clear the regular expression caches"
237    _cache.clear()
238    _compile_repl.cache_clear()
239
240def template(pattern, flags=0):
241    "Compile a template pattern, returning a pattern object"
242    return _compile(pattern, flags|T)
243
244_alphanum_str = frozenset(
245    "_abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ01234567890")
246_alphanum_bytes = frozenset(
247    b"_abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ01234567890")
248
249def escape(pattern):
250    """
251    Escape all the characters in pattern except ASCII letters, numbers and '_'.
252    """
253    if isinstance(pattern, str):
254        alphanum = _alphanum_str
255        s = list(pattern)
256        for i, c in enumerate(pattern):
257            if c not in alphanum:
258                if c == "\000":
259                    s[i] = "\\000"
260                else:
261                    s[i] = "\\" + c
262        return "".join(s)
263    else:
264        alphanum = _alphanum_bytes
265        s = []
266        esc = ord(b"\\")
267        for c in pattern:
268            if c in alphanum:
269                s.append(c)
270            else:
271                if c == 0:
272                    s.extend(b"\\000")
273                else:
274                    s.append(esc)
275                    s.append(c)
276        return bytes(s)
277
278# --------------------------------------------------------------------
279# internals
280
281_cache = {}
282
283_pattern_type = type(sre_compile.compile("", 0))
284
285_MAXCACHE = 512
286def _compile(pattern, flags):
287    # internal: compile pattern
288    try:
289        p, loc = _cache[type(pattern), pattern, flags]
290        if loc is None or loc == _locale.setlocale(_locale.LC_CTYPE):
291            return p
292    except KeyError:
293        pass
294    if isinstance(pattern, _pattern_type):
295        if flags:
296            raise ValueError(
297                "cannot process flags argument with a compiled pattern")
298        return pattern
299    if not sre_compile.isstring(pattern):
300        raise TypeError("first argument must be string or compiled pattern")
301    p = sre_compile.compile(pattern, flags)
302    if not (flags & DEBUG):
303        if len(_cache) >= _MAXCACHE:
304            _cache.clear()
305        if p.flags & LOCALE:
306            if not _locale:
307                return p
308            loc = _locale.setlocale(_locale.LC_CTYPE)
309        else:
310            loc = None
311        _cache[type(pattern), pattern, flags] = p, loc
312    return p
313
314@functools.lru_cache(_MAXCACHE)
315def _compile_repl(repl, pattern):
316    # internal: compile replacement pattern
317    return sre_parse.parse_template(repl, pattern)
318
319def _expand(pattern, match, template):
320    # internal: match.expand implementation hook
321    template = sre_parse.parse_template(template, pattern)
322    return sre_parse.expand_template(template, match)
323
324def _subx(pattern, template):
325    # internal: pattern.sub/subn implementation helper
326    template = _compile_repl(template, pattern)
327    if not template[0] and len(template[1]) == 1:
328        # literal replacement
329        return template[1][0]
330    def filter(match, template=template):
331        return sre_parse.expand_template(template, match)
332    return filter
333
334# register myself for pickling
335
336import copyreg
337
338def _pickle(p):
339    return _compile, (p.pattern, p.flags)
340
341copyreg.pickle(_pattern_type, _pickle, _compile)
342
343# --------------------------------------------------------------------
344# experimental stuff (see python-dev discussions for details)
345
346class Scanner:
347    def __init__(self, lexicon, flags=0):
348        from sre_constants import BRANCH, SUBPATTERN
349        self.lexicon = lexicon
350        # combine phrases into a compound pattern
351        p = []
352        s = sre_parse.Pattern()
353        s.flags = flags
354        for phrase, action in lexicon:
355            gid = s.opengroup()
356            p.append(sre_parse.SubPattern(s, [
357                (SUBPATTERN, (gid, 0, 0, sre_parse.parse(phrase, flags))),
358                ]))
359            s.closegroup(gid, p[-1])
360        p = sre_parse.SubPattern(s, [(BRANCH, (None, p))])
361        self.scanner = sre_compile.compile(p)
362    def scan(self, string):
363        result = []
364        append = result.append
365        match = self.scanner.scanner(string).match
366        i = 0
367        while True:
368            m = match()
369            if not m:
370                break
371            j = m.end()
372            if i == j:
373                break
374            action = self.lexicon[m.lastindex-1][1]
375            if callable(action):
376                self.match = m
377                action = action(self, m.group())
378            if action is not None:
379                append(action)
380            i = j
381        return result, string[i:]
382